Loading Events

« All Events

  • This event has passed.

MLOps for Data Scientists

September 20 @ 11:50 am - 1:00 pm

Please register using the zoom link to get a reminder:


Data scientists must build model training pipelines instead of providing trained models to operation engineers in order to implement an efficient MLOps workflow. In order to create such pipelines within the Jupyter notebook environment, we developed a JupyterLab extension. By using this extension, data scientists can create seamless model training pipelines with minimal changes to their workflow.

– We describe how we designed and implemented a free-of-charge pipeline runtime for JupyterLab. The extension is called Link. This extension provides pipeline creation and execution runtimes. There has been little change to the original behavior of JupyterLab. As a result, most data scientists can create pipelines in JupyterLab in just a few minutes.

– Version control is necessary for pipeline codes. Each pipeline code is contained in a .ipynb file. Consequently, we implemented another JupyterLab extension for git support called Link-git. The extension provides git drivers for Jupyter notebooks with the Link pipeline specification. The design and implementation will be explained in detail.

– It is possible to export and integrate pipelines with open source MLOps platforms like Kubeflow with a single click. Nevertheless, MLOps platforms must provide support for Link extensions in order to fully utilize MLOps capabilities such as automatic execution environment snapshots, pipeline debugging, model performance comparison, and model deployment.


11:50 am – 11:55 am Arrival and socializing and Opening 11:55 am – 1:00 pm Sangwoo Shim, “MLOps for Data Scientists” 1:00 pm – 1:10 pm Q&A

About Sangwoo Shim:

Sangwoo Shim is Chief Technology Officer (CTO) and co-founder at MakinaRocks. He focuses on developing Machine Learning infrastructure and overseeing AI projects and solutions for manufacturing. Sangwoo has pursued his career in various quantitative areas including quantitative finance, equity trading, and data analysis in global companies. Upon completion of his Ph.D. in Chemical Physics at Harvard University, he joined Bank of America Merrill Lynch in New York as a quantitative analyst. He later worked as a quantitative portfolio manager at WorldQuant and Millennium Capital in Old Greenwich and Singapore. Prior to MakinaRocks, he worked in the Big Data Analysis Group at Samsung Electronics. Sangwoo believes that his expertise in statistics, optimization, and software engineering will contribute to achieving MakinaRocks’s mission to innovate manufacturing through artificial intelligence.

Please register using the zoom link to get a reminder:


Webinar ID: 879 1731 5264


September 20
11:50 am - 1:00 pm